EA://INTEL — A structural check for meaning without counting primitives

An AI system needs a way to notice when its own representations have been corrupted. It cannot check itself against itself because that is circular. It can check itself against a fixed mathematical structure it did not write and rebuild from published rules. That check is called

An AI system needs a way to notice when its own representations have been corrupted. It cannot check itself against itself because that is circular. It can check itself against a fixed mathematical structure it did not write and rebuild from published rules. That check is called a comparator, that is a routine that reports any difference between two structures. The checking layer enforces local structural validity under stated invariants — properties a structure is required to have, fixed in advance. It does not make a system good because a system could pass every check while pursuing anything harmful. Much of the last year was spent finding that earlier results did not hold and recording why those failures occurred.

The project built four things that are ready for inspection. First, a public register lists every claim with its standing and evidence. Second, an append-only log records the programme’s own failures. Third, the mechanical gate refuses text that breaks the rules. Fourth, mathematical results verified by running code show exact correspondence between one framework structure and an object from mathematical physics. The result that would let any expressive language be embedded faithfully has a located gap — a specific identified step in an argument that does not go through. Its argument needs the source to be free of internal identities because real languages have synonymy. Repair over a non-convex space of acceptable states is provably multivalued, so the operation returns several answers instead of one.

An analogy: headline counts of proven results appeared in materials for months until an audit found they grew by accumulation rather than derivation. That number was withdrawn and now appears only in the failure log recording its own retirement. The strongest objection is that this involves a great deal of open question for a programme asking others to take it seriously. Agreed, and that is the reason for publishing the register rather than a description of settled theory. The open questions are the work; a programme with none would either be finished or not looking.

A reader who has read nothing else must understand what is established versus open without guessing. The claim about existence carries its own standing in the register and should not be taken from this summary alone. Which of those open questions would you attack first, and what would you need from us to try?


Write a comment